Introduction
Equitable clinical evaluations are essential in distributed training
models, where identical competencies must be assessed consistently regardless
of campus or clinical site. At Texas A&M University College of Medicine,
clinical evaluations comprise half of clerkship grading and thus
substantially influence student ranking, a metric medical schools use to
differentiate performance in competitive application cycles. Because student
rank directly impacts outcomes such as residency selection and interview
visibility, any systematic scoring variability between campuses may create
unintended inequities. These discrepancies carry downstream implications for
matching into residencies where program directors often incorporate medical
school performance markers when evaluating applicants. Ensuring fairness in
evaluations therefore supports both assessment validity and equal opportunity
for residency advancement within geographically distributed medical education
environments.
Methods
Clinical-evaluation scores from five core campuses (Houston, Willowbrook,
Dallas, Round Rock, and Bryan/College Station (BCS)) were analyzed across six
required clerkships (Family Medicine, Internal Medicine, Pediatrics, Psychiatry,
OB/GYN, and Surgery). Cohorts corresponded to Classes of 2024-2027. One-way
ANOVA tested campus influence on clinical evaluation; Tukey HSD post-hoc
identified pairwise campus differences (?=0.05, FWER-corrected).
Results
Significant campus-level variabilities were found across multiple
clerkships and years. Our pooled data set (n=3,390) demonstrated campus
significantly affected clinical evaluation scores (ANOVA p<10?¹?),
explaining 2.2% of overall score variance (?² = 0.022). Post-hoc found BCS
underperforming compared to Dallas and Round Rock, Willowbrook outperforming
compared to Dallas and Houston, and Houston underperforming compared to
Dallas and Round Rock.
Discussion
Evaluation scoring differs collectively and pairwise across multiple
distributed campuses, with direction and magnitude dependent on clerkship,
suggesting rotation-level evaluator norms or learning-environment influences.
Quantifying these differences supports targeted faculty development and
fairer assessment calibration in distributed medical-education models.